mastodon.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
The original server operated by the Mastodon gGmbH non-profit

Administered by:

Server stats:

377K
active users

Allison Parrish

so, question: let's say I have a big list of log probability estimates for English unigrams (from spacy) and a small text file (say <500 words). is there a reasonable and well-motivated way to extract keywords/topics from the small text file using *only* the frequency of tokens in the text file and the tokens' corresponding unigram probabilities? (tfidf etc won't work I don't think because I'm not really comparing documents)